pelizzai-team

Coordinate multiple AI agents for parallel tasks with adversarial verification.

Updated Jun 26, 2026

Search Tags:#workflow #delegation #multi-agent #coordination #engineering #parallelism

Ready to run

npx skills add https://github.com/rpelizza/PelizzAI --skill pelizzai-team

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: pelizzai-team
Source: https://github.com/rpelizza/PelizzAI/tree/main/.agents/skills/pelizzai-team
Command: npx skills add https://github.com/rpelizza/PelizzAI --skill pelizzai-team

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the challenge of coordinating multiple AI agents to work in parallel on complex, multi-faceted projects, ensuring that tasks are decomposed, delegated, and synthesized without the overhead of manual management.

Core Features & Use Cases

Parallel Coordination: Manages multiple agents working on distinct fronts, such as cross-layer development or competing research hypotheses.
Dual-Mode Execution: Automatically switches between native Claude Code Agent Teams and a robust subagent fallback system.
Adversarial Verification: Implements cross-checking and refutation rounds to ensure high-quality, verified deliverables.

Quick Start

Use the pelizzai-team skill to coordinate a team of agents to investigate and fix the reported performance regression in the backend module.

Frequently Asked Questions about pelizzai-team

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I coordinate multiple AI agents for parallel engineering tasks?▼

Multi-agent coordination is achieved through structured task decomposition, role-based delegation, and disjoint file management, allowing agents to execute distinct engineering fronts in parallel. The system then synthesizes deliverables via adversarial verification.

What is adversarial verification in multi-agent team coordination?▼

Adversarial verification is a cross-checking and refutation process applied after parallel agents complete their tasks. It ensures high-quality deliverables by having agents review and refute each other's work before final synthesized delivery.

How do I delegate complex engineering projects across cross-layer implementations?▼

Complex engineering projects are delegated by decomposing them into disjoint file management tasks assigned to specialized agents. This supports cross-layer implementation by ensuring agents work on distinct fronts without overlapping.

Can I use multi-agent coordination for competing hypothesis investigation?▼

Yes, multi-agent coordination supports competing hypothesis investigation by assigning agents to explore different research fronts in parallel. The system synthesizes findings through structured adversarial verification rounds.

Does the multi-agent team skill require native Claude Code Agent Teams to function?▼

No, the skill features dual-mode execution that automatically switches between native Claude Code Agent Teams and a robust subagent fallback system, ensuring multi-agent coordination works across different environments.

When should I avoid using parallel multi-agent coordination for engineering tasks?▼

Parallel multi-agent coordination is not suited for simple, single-threaded tasks that lack multi-perspective review needs or cross-layer implementation requirements, as the overhead of decomposition and adversarial verification outweighs benefits.